Dynamic monitoring and characteristic analysis of a long-span operational bridge from high-rate sensor responses using RAAVMD approach

To investigate high-rate sensors in identifying structural characteristics of an operational long-span cable-stayed bridge, the 50 Hz GNSS-RTK and 100 Hz accelerometer were employed for dynamic monitoring. Firstly, considering the background noise influence, a GNSS test in different environments is conducted to analyze noise characteristics and GNSS performance. Then, a new improved variational mode decomposition (defined as RAAVMD) is developed to extract dynamic responses of the bridge by alleviating noise. The evaluation results show that the proposed method outperforms the other four methods in noise reduction of signals with different noise levels. The experimental results of the cable-stayed bridge demonstrate that the structural failure probability obtained from 50 Hz GNSS-RTK semi-static displacement is 0.0578; RAAVMD effectively reduces the noise in the measuring responses of high-rate sensors; the frequencies and their variation characteristics in different operating periods are effectively identified from high-rate GNSS-RTK and acceleration responses based on the proposed approach.